Fast Fault Diagnosis for Proton Exchange Membrane Fuel Cells based on an Electrochemical Impedance Spectroscopy Measurement

Abstract

The proton exchange membrane fuel cell (PEMFC) is considered as one of the most promising energy sources. For the large commercialization, it is necessary to improve the reliability and durability of PEMFCs with the fault diagnosis approach. This paper proposes an online fault diagnosis method for PEMFCs based on fast electrochemical impedance spectroscopy (EIS) measurement. Specifically, the periodic triangular wave is utilized as a disturbance current, and the PEMFC’s impedance spectrum is obtained through a fast Fourier transform (FFT). Then, the parameters of an improved equivalent circuit model are identified based on the nonlinear least squares method. After that, the support vector machine (SVM) with a one-vs-all strategy is established to diagnose four operation states. Experimental results show that the proposed fast fault diagnosis method can detect different faults accurately with strong generalization and robustness.